© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 1651
    Prev
    Next
    INDEX
    Explanations

    programming keywords

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron strongly activates on single‐word code identifiers that are function or API method names (e.g. Log, Trace, Store, list, time, Query, lock, cache, Hash).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     estando
    -1.59
    內容簡介
    -1.55
     elevado
    -1.52
     новых
    -1.52
     nachdem
    -1.46
    太大
    -1.46
     снова
    -1.45
    在其
    -1.44
     continually
    -1.41
     finds
    -1.40
    POSITIVE LOGITS
     our
    2.23
    ;
    2.13
     not
    2.06
     them
    2.05
     one
    1.94
     for
    1.93
     összefoglal
    1.89
     your
    1.79
     four
    1.77
     six
    1.74
    Activations Density 0.074%

    No Known Activations